Professional Documents
Culture Documents
A R T I C LE I N FO A B S T R A C T
Keywords: A combination of mechanical methods was developed to obtain a suspension of nanocellulose (NC) from
Mechanical methods Eucalyptus sawdust. In this two-step process, the ball milling (1–4 h) was followed by a high-intensity ultrasound
Cellulose nanostructures (10–30 min) irradiation. The prepared samples were characterized using Fourier-transform infrared spectro-
Ball mill scopy, Raman, dynamic and static light scattering, Zeta potential, X-ray diffraction, X-ray photoelectron spec-
High-intensity ultrasound
troscopy, Thermogravimetric analysis, Field-emission scanning electron microscopy, and Transmission electron
Eucalyptus sawdust
microscopy. The crystallinity was calculated using different methods and compared. The mechanical grinding
Nanoparticles’ characterization
results in a random and irregular break of the fibers. In a combined method, the ultrasound waves generate more
defibrillation and separation of the nanofibers and, consequently, the reduction of fiber diameters. The high-
intensity ultrasound promoted an increase in the homogeneity, crystallinity, and electrostatic and thermal sta-
bility of the NC suspension. The samples milled for 2 h and 20 min ultrasound irradiation presented the high-
efficiency in obtaining NC.
1. Introduction modulus, unique morphology, and excellent optical properties [5]. Due
to the unique properties of NC, this material is applicable in many
Cellulose is an environmentally friendly biopolymer derived from fields, such as mechanical reinforcements in composite materials,
natural sources such as forest waste, woody biomass, crops, algae, and bioengineering, pharmaceutical, and biomedical technologies [6]. A
microorganisms. Cellulose and its derivatives are sustainable and bio- wide range of applications and products can be developed; therefore,
degradable, therefore receiving exceptional attention as a substitute for NC is a fascinating next-generation engineering material to researchers
petrochemical feedstock in polymer industries [1]. More recently, uses and industries.
of cellulose from the forest residues have been devoted to the isolation Several top-down approaches are available for NC isolation, such as
of nanocellulose (NC), and Eucalyptus sawdust is one of them. Eu- chemical, biological, mechanical, and combination of chemical and
calyptus is a fast-growing wood that covers a larger planted area with mechanical [7–9]. However, an environmentally-friendly combination
approximately 187 million ha in the world [2]. During the processing of method that does not employ strong acids or toxic reagents has not been
eucalyptus, an enormous amount of sawdust is originated, which is rich widely explored. Various chemical processes are required to transform
in lignocellulosic constituents ˜25−35 wt.% of lignin, ˜20−30 wt.% of the lignocellulose rich sawdust into cellulose pulp [10].
hemicellulose and ˜35−40 wt.% of cellulose [3]. The residues are often Usually, chemical methods are conducted by acid hydrolysis using
disposing as waste or burning for energy production. It results in high strong acids, and it is well established [5,11–14]. However, chemicals
emissions of pollutant gas [4] leads to undesirable environmental are always risky to handle, corrosive to the equipment’s, and detri-
consequences. mental effect on the environment [5,10]. The enzymatic hydrolysis
Nanocellulose is a new class of eco-material with unique properties represents the biological method. This method requires long reaction
such as nanoscale dimension, high surface area, high strength, specific times, and the cost of enzymes is usually high and also limited uses
⁎
Corresponding author.
E-mail addresses: reis.f@ufabc.edu.br (R.R. Ferreira), alana.gabrieli@ufabc.edu.br (A.G. Souza), lucas.nunes11@fatec.sp.gov.br (L.L. Nunes),
nshahi1164@tuskegee.edu (N. Shahi), vrangari@tuskegee.edu (V.K. Rangari), dervalrosa@yahoo.com.br (D.d.S. Rosa).
https://doi.org/10.1016/j.mtcomm.2019.100755
Received 3 August 2019; Received in revised form 28 October 2019; Accepted 5 November 2019
Available online 07 November 2019
2352-4928/ © 2019 Elsevier Ltd. All rights reserved.
R.R. Ferreira, et al. Materials Today Communications 22 (2020) 100755
[15,16]. On the other hand, mechanical methods are simple and do not solution of NaOH (10 wt.%) and KOH (10 wt.%) for 2 h under magnetic
require chemical reagents; this type of method are outstanding due to stirring, washed and dried at 50 °C for 12 h. After these treatments, the
their high production capacity and low cost [17–19]. The main ex- final material has in its composition ˜1.5 wt.% of hemicellulose and
amples of mechanical methods are ball mill, cryogenic break, and ul- ˜2 wt.% of lignin, and this sample is named T2. This methodology was
trasound. previously reported [30]. The yield was ˜40 wt.%. Considering that the
The ball mill is an environmentally friendly and low-cost technique eucalyptus residues have ˜40 wt.% of cellulosic fibers and ˜60 wt.% of
that is widely applied in the industry all over the world. It consists of a other non-cellulosic components, this yield is considered as excellent.
cylindrical shell rotating around its axis, which is partially filled with
balls; the energy released from impact and attriction between the balls 2.2.2. Isolation of nanocellulose
and the samples converts the particles from micro- to nanoscale [20]. The treated sample, after the second treatment, was mixed with
However, some disadvantages include the possibility of contamination, ethanol (80 % v/v) in the proportion 1:1 g/ml and subjected to 4 dif-
usually ceramic or metal, from grinding medium and container, the ferent milling times: 1 h, 2 h, 3 h, and 4 h, in a ball mill (MA500 -
formation of irregular shapes, and extended milling and cleaning times Marconi). This is rotational type mill, and the particles are shrunk by
[5,21]. The current literature survey shows the frequent uses of pla- frictional and friction forces that occur between the balls, the jar wall,
netary ball milling (high-energy ball mill) or shaker mill for obtaining and the sample during the rotational movement of the jar. The speed of
NC, and/or acid solutions during the milling. The planetary ball mill the milling is 200 rpm. The jars used are made of porcelain, model 21A
achieves higher rotations and requires much energy; it has very low (Chiarotti), with the following characteristics: diameter of 150 mm,
milling capacity due to the small containers and is not common on an height of 195 mm, capacity of 1 L. The jar had 70 % of its volume filled
industrial scale [5,17,22]. The tumbler ball mill, which was used in this with balls and 30 % with samples. The jar was filled with 5 g of treated
work, is a very simple process that depends mainly on the time and sample, 5 ml of ethanol, 305 g of larger alumina balls (diameter of
balls and diameter of the container [21,23,24]. 21 mm), and 44 g of smaller alumina balls (12 mm). The balls are 92 %
While the ball mill involves friction and shear forces, the ultrasound pure, the other 8 % being silica or silicon dioxide. The ball mill con-
process applies the use of oscillating power to create hydrodynamic ditions were selected according to our previous studies [20,31]. A
forces of ultrasound [25]. The high intensive waves result in the for- photograph of the jar and balls used is presented in Fig. 1 S (Supple-
mation, expansion, and implosion of microscopic gas bubbles when the mentary material). The production yield of the ball-milled product is
sample absorbs the energy, and this intense energy decreases the size of ˜80 wt.% (this value was calculated comparing the sample added in the
the particles [26,27]. However, generally high times or low sample jar of ball mill and the sample obtained after the milling time).
quantity per process need to be used. Aiming to find an environmental After 1 h interval (each hour) of milling, the samples were oven-
process with high efficiency and low cost, researches are combining two dried at 35 °C for 24 h to remove the ethanol. Then the samples in
or more isolation methods at the same time or in sequence. The com- powder form were subjected to high-intensity ultrasonic treatment in a
bination of mechanical methods, such as ball mill and high-intensity Sonics Vibra Cell, at the power of 400 W and 24 kHz, in a continuous
ultrasound have recently been highlighted [22]. process. During the ultrasonication, the samples were maintained under
The purpose of the study was to develop a combined mechanical an ice bath. The tip diameter is 14 mm, with an amplitude of 20 %. For
method (ball milled and ultrasound) to isolated NC from Eucalyptus each sample, aqueous solutions containing 1 wt.% of NC were sonicated
sawdust. This method can be extended to large scale production, which during the selected time: 10, 20, and 30 min. After the ultrasound, the
is different from the existing literature, is simple, low cost, and com- samples were in suspension form (1 wt.%). The image presented in
mercially attractive. It is important to mention that the production of Fig. 1 represents the followed methodology in this study, and Table 1
nanocellulose in a pilot-scale is currently carried out mainly by acid shows the adopted nomenclatures.
hydrolysis [28,29]. The NCs isolated were characterized by morpho-
logical, chemical, physical, and thermal properties. Furthermore, the
2.2.3. Characterization
crystallinity of the NC was compared with XRD, Raman, and FTIR.
2.2.3.1. Fourier transform infrared spectroscopy. Fourier transform
Therefore, this study aimed to transform Eucalyptus residues as pro-
infrared spectroscopy (FT-IR) was conducted using a Frontier 94,942
mising sources of high value-added biopolymer (in the NC form).
(PerkinElmer, USA) in the range of 4000 to 400 cm-1 with 32 scans and
a resolution of 4 cm-1 with attenuated total reflectance (ATR) accessory
2. Experimental
equipped with diamond. The FTIR relative crystallinity (index of
crystallinity - IC) was calculated using the ratio of the absorption
2.1. Materials
bands at 1369 and 1315 cm-1, as reported by Lee et al. (2015) [32].
Pellet pressed samples were analyzed at room temperature.
Eucalyptus citriodora waste (eucalyptus sawdust – ES) was kindly
donated after harvesting and cutting at the state of Mato Grosso, Brazil.
The composition of the eucalyptus sawdust was ˜33 wt.% of lignin, 2.2.3.2. Raman spectroscopy. All samples were analyzed with a Raman
˜27 wt.% of hemicellulose, and ˜40 wt.% of cellulose. All the reagents spectroscopy (XRD Raman Spectroscopy Thermo Scientific, Madison,
were purchased from Sigma-Aldrich (SP-Brazil) (NaClO2, 99 %, NaOH USA). This system is equipped with a 780 nm wavelength with 5.0 mW
-95 %, KOH-98 %, ethanol–98 %). power laser. Approximately 0.5 mm–1 mm length of each dried sample
is analyzed at 50x magnification. The spectra were recorded, and each
2.2. Methods spectrum presented is an average of 3 different points on the sample.
The data acquisition was carried out in a range of 100-3500 cm−1, and
2.2.1. Treatments 32 scans were accumulated in each sample with 4 cm-1 spectral
To obtain cellulose without the presence of lignin and hemi- resolution. All the graphs were baseline corrected (25 points) using
cellulose, two chemical pre-treatments were carried out in sequence: OriginPro 2009. From the subtracted spectrum of the NCs, peak
intensities for the 380 and 1096 cm-1 bands were measured. The
2.2.1.1. First treatment. The ES was first treated in a solution of NaClO2 intensity ratio I380/I1096 was determined, and Raman crystallinity
(3.8 wt.%) for 2 h at 50 °C under mechanical stirring, and then the (index of crystallinity – IC) was calculated using Eq. 1 [32–34].
obtained material was washed and dried in an oven at 50 °C for 12 h.
[(I380 − I1096)) − 0.0286]
IC =
0.0065 (1)
2.2.1.2. Second treatment. Subsequently, the material was treated in a
2
R.R. Ferreira, et al. Materials Today Communications 22 (2020) 100755
Fig. 1. Flow diagram representative of the preparation of nanocellulose: the pretreatment methods, followed by the mechanical grinding (ball mill) and finishing
with assisted ultrasound methodology.
2.2.3.3. Dynamic and static light scattering. The isolated NC 2.2.3.3.1. Morphological analysis – field emission scanning electron
measurements were performed using dynamic light scattering (DLS) microscopy and transmission electron microscopy (FE-SEM and TEM). The
with a stable 90° scattering angle. Static light scattering (SLS) was surface morphology of all the samples was analyzed at different stages
conducted in the region 30–150 °. Both samples were performed using of treatments using FE-SEM (JEOL-JSM-7200 F) microscope equipped
the ALV/CGS-3 instrument consisting of a polarized HeNe laser with energy-dispersive X-ray spectroscopy (EDS), with an acceleration
(22 mW). The equipment was operating at a wavelength (ʎ) 633 nm, voltage of 10 kV. The sample was prepared; drop of the NCs (2 wt.%)
with NC diluted to a concentration of 0.01 wt%. The reading time was dropped in the stub and after dried in the stove at 60 °C during 2 h.
established to obtain the DLS data was 10 s, and the measurements were or Samples were mounted on carbon conductive tape and sputtered for
performed in triplicate using the final average values. Hydrodynamic around 5 min in Hummer 6.2 sputter system under 10 mA.
radius (RH) was calculated by using Eq. 2 [35]: The microstructure and morphology of the samples with the best
conditions were investigated by Transmission Electron Microscope
KB T
RH = (TEM-Joel 2010). A drop of the colloidal solution at 2 wt.% was de-
3πηD (2)
posited on a copper grid for analysis.
Where: KB is the Boltzmann's constant, T is the temperature, η is the
viscosity, and D is the translational diffusion coefficient. 2.2.3.4. Zeta potential. The zeta potential measurement was
In the SLS experiment, the angular dependence of the absolute ex- determined using Zetasizer Nano-ZS (Malvern Instruments) to
cess time-averaging (for a duration of scanning of 10 s) scattering in- evaluate electrophoretic mobility of particles converted to Zeta
tensity, known as the Rayleigh ratio Rvv (θ) leads to the mean square potential using the Smoluchowski [30] Eq. 3. The reading time
radius of gyration (RG), which allows identification of the geometric established to obtain the zeta potential data was 10 s, and the
shape of the objects (RG/RH) [36]. measurements were performed in triplicate using the final average
Table 1
Nomenclatures of samples prepared by varying milling and high-intensity ultrasonic times.
SAMPLE NOMENCLATURE SAMPLE NOMENCLATURE
1M 1h milling 3M 3h milling
1 M-10U 1h milling - 10 min ultrasound 3 M-10U 3h milling - 10 min ultrasound
1 M-20U 1h milling - 20 min ultrasound 3 M-20U 3h milling - 20 min ultrasound
1 M-30U 1h milling - 30 min ultrasound 3 M-30U 3h milling - 30 min ultrasound
2M 2h milling 4M 4h milling
2 M-10U 2h milling - 10 min ultrasound 4 M-10U 4h milling - 10 min ultrasound
2 M-20U 2h milling - 20 min ultrasound 4 M-20U 4h milling - 20 min ultrasound
2 M-30U 2h milling - 30 min ultrasound 4 M-30U 4h milling - 30 min ultrasound
3
R.R. Ferreira, et al. Materials Today Communications 22 (2020) 100755
values. The NCs were diluted to a concentration of 0.01 % wt.%. This value is not the real value and was used only for comparing the
samples with each other [32].
3 η UE
ζ= The ball mill is a challenge composed of three main aspects: (i)
2 ε f (ka) (3)
energy and time, (ii) size and stability of nanoparticles, and (iii) wear:
contamination and amorphization (Fig. 3).
In general, process optimization needs to balance the three chal-
2.2.3.5. X-ray diffraction. X-ray powder diffraction data were collected
lenges for shorter times and energy, smaller particle sizes, high stabi-
on a D8 Focus diffractometer (Bruker AXS - Karlsruhe, Germany),
lity, and crystallinity. According to the crystallinity values, it is not
operating at 40 kV and 40 mA, with monochromatic CuKα1 radiation
possible to unite all the desired characteristics in one material only with
(λ =1.54056 Å) selected by a curved Ge (111) crystal monochromator.
the ball mill: the sample 2 M showed high crystallinity (however, as
The data were collected from 10° to 60° in steps of 0.01° and a counting
presented in the next sections, the sizes were larger than the desire for
time of 100 s at each 0.5°. The index of crystallinity (IC) was calculated
this type of nanoparticles).
according to the following equation (Eq. (4)):
Less crystallinity of the cellulose obtained from samples 1 M, 3 M
IC(%)= (I200-Iam)/I200 *100 (4) and 4 M are justified by the inefficient break resulting in the impact and
shear forces or to the excessive time that resulted in the breaking of
Where I200 is referred to the maximum intensity of the 200 lattice dif-
both the amorphous and crystalline regions. Combine mechanical
fraction peak at 2θ = 22.2° and Iam is the intensity of diffraction for the
method is an alternative to use low times in the ball mill and guarantee
amorphous part at 2θ = 18.5° [37,38].
the aimed properties in the dimensional and stability characteristics of
the nanocellulose.
2.2.3.6. X-ray photoelectron spectroscopy. The XPS analysis was The comparison between the FTIR spectra of the milled samples and
conducted on a ThermoFisher Scientific, model K-alpha+, equipped its ultrasonicated samples is available in Fig. 2S on the supporting
with a monochromatic A1 Kα radiation at 1486.6 eV. The analysis was material. All samples showed slight increases in the IC since ultrasonic
performed at room temperature and operated at 10 or 20 eV for high- treatment acts in the separation of the nanoparticles. Longer times may
resolution scans. Data were treated with CASAXPS software. have resulted in an entanglement of the fibers, which resulted in dis-
turbance of the crystalline region. The FT-IR spectra analysis also al-
2.2.3.7. Thermogravimetric analysis. Thermogravimetric analysis (TGA) lows a pre-selection of the samples based on the crystallinity: 2 h of
was carried out using STA 6000 instrument (PerkinElmer, USA). The milling and 20 min of ultrasonication.
samples were heated from 20 to 600 °C with a constant heating rate of Similar to observed in the FTIR analysis, Raman frequency, in-
20 °C/min under nitrogen flow. All the graphs were analyzed using the tensity, and band shape of the vibrations may vary between the nano-
software OriginPro 2009. celluloses due to the differences in the chemical structure, interaction
between the biopolymer chains, and its organization. In most cases, the
3. Results and discussion peak sharpness and intensities are associated with the crystallinity of
NC in which, increase in peak intensities increases in the crystallinity of
3.1. Chemical characterization of the eucalyptus biomass and its NC as shown in Fig. 4a–b. Crystallinity percentage after ultrasonic ir-
nanocelluloses radiations drastically increase in all the study samples, as presented in
Table 2; this may be due to the disruption of amorphous domains to
A basic characterization of the chemical structure of the samples yield nanocellulose because of ultrasonic irradiation.
with different grinding and ultrasound times was conducted for a pre- Raman spectra of cellulose from eight different treatments were
vious evaluation of the fiber composition and apparent crystallinity collected and compared to the literature data. Cellulose is a poly-
using FTIR and Raman spectroscopy. Fig. 2a presents the full spectra of saccharide thus has characteristics Raman bands from 300 to 3000 cm-
1
the milled samples. . The main cellulosic peaks are indicated with arrows, and corre-
All the samples showed similar spectra, which is indicative of the sponding peaks are ˜380 and ˜460 cm-1: CeCeC, C–O, CeC–O ring
similarity between the chemical composition of the samples [22]. It was deformation. The band at 900 cm-1 is associated with the C–H ring of
observed bands of 3500-3200 and 2900 cm-1, characteristic of the −OH cellulose and is the disorder of the cellulosic structure (amorphous re-
binding and C–H symmetrical stretching groups of cellulose [31,39]. gion) [33]. Similarly, 1340 cm-1: −CH2, HeCeC, HeOeCe, and
The bands around 898, 1030, 1160, 1315, 1369, and 1425 cm-1 were C–OeH; 1456 cm-1: HeC and HeOeC bending. Raman bands from
assigned to the C–H vibrations, C–O stretching, C–O-C asymmetric vi- 1300 to 1500 cm-1 corresponding to −CH2 and −CH2OH deformations
bration, −CH2 rocking vibration and −CH2 deformation vibration of are also reported [33]. The Raman bands at 2889 cm-1: CH and CH2
the cellulose, respectively [40]. corresponding to stretching and bands at 3286−3402 cm-1: OeH
Fig. 2a also shows a zoom of the region between 1800 and 1400 cm- stretch [42].
1
; it is evident that all the milled samples do not show the peaks related Raman technique can result in estimate values of crystallinity, as
to the lignin and hemicellulose [40,41]. Other peaks assigned to the presented in Fig. 4a–b. This value can be influenced by hemicelluloses
cellulose structure are found at 1155, 1060, 1010, 990, and 893 cm-1. (as impurities), and therefore other measurement methods need to be
Most of these peaks are influenced by neighboring bonds and their vi- conducted and compared. of crystallinity by the ratio of cellulose's
brational modes, which results in high band overlap and shift of some Raman bands at 380 and 1096 cm-1. Also, the region of the sample or
peaks, especially below 1100 cm-1. After the milling, it was noted a the preparation of the sample can influence the obtained values, and
significant decrease in the intensity of the band 1640 and this is the reason why the results of crystallinity measurements by
3400−3000 cm-1 due to the increase in the apparent crystallinity of the different researches have been, at times, contradictory [34].
samples; increasing chain ordering, there is less available spacing for It was possible to sequence the crystallinity for the milled samples as
water entering the sample. Functional groups between 1300 to 800 cm-1 2 M ˜ 3 M > 1 M > 4 M; the results indicate that the best condition of
were observed; the increasing intensity of these groups indicates in- milling time considering only the crystallinity of the samples is two
creased crystallinity of cellulose in the milled samples. hours. Evaluating the ultrasonicated samples, all of them showed sig-
The apparent index of crystallinity (IC) was calculated for the nificant increases in the crystallinity values. The defibrillation effect
samples 1 M, 2 M, 3 M, and 4 M. The sample ES showed an overlap of can justify the increase in the IC values after the high-intensity ultra-
peaks due to the presence of non-cellulosic compounds, and it was not sound: when the ultrasonic waves arrive at the sample, they penetrate
possible to calculate the "apparent" IC, as presented in Table 2 [16,32]. the crystalline structure of the cellulose and result in the separation of
4
R.R. Ferreira, et al. Materials Today Communications 22 (2020) 100755
Fig. 2. Infrared spectra of (a) Eucalyptus (ES) and 1 M, 2 M, 3 M and 4 M samples and zoom in the region 1800 to 1400 cm-1 and 1390 to 600 cm-1; and (b) full spectra
for the ultrasound-20 min samples.
Table 2 this technique is the hydrodynamic radius (RH) of a sphere having the
Crystallinity values obtained from FTIR and Raman spectra. same diffusion coefficient as the rod-like cellulose nanoparticles
Samples Apparent IC (%) (FTIR) 1369/1315 Crystallinity (%) (Raman)
[8,43,44]. Fig. 3S shows the curves of the size distribution and in-
formation about the maximum, minimum, and average sizes of the
1M 98.4 49.8 milled samples.
1 M-20U 99.3 57.9 Comparing the milled samples, in Fig. 3S, it was possible to observe
2M 99.4 59.9
2 M-20U 99.7 77.0
that the RH Average was the lowest for 2 h of ball mill, followed for 3 h,
3M 88.1 60.3 4 h, and 1 h. The milling time of the 1 h was not enough to break the
3 M-20U 89.9 73.5 fibers into nanostructures; the higher times (3 and 4 h) show a trend of
4M 98.1 17.7 increase de dimensions. The best milling time was 2 h. According to
4 M-20U 99.9 66.3
Sofla et al., longer grinding times may result in the growth of a fibrous
layer of cellulose around the grinding balls. This fact decreases the
impact power of the balls on the particles and results in larger sizes
the nanofibrils. This separation, associated with the high shear force
[45,46]. This behavior justifies the larger particle sizes of the samples
present in the solution, breaks the amorphous regions and isolates the
3 M and 4 M.
crystalline ones, resulting in more separated and crystalline NCs. The
In all DLS measures, it was verified that the sample with 20 min of
more crystalline the sample ground, the easier the separation of na-
ultrasound showed the lowest RH averages. Most of the particles also
nofibrils, and therefore the best results among the ultrasonic samples
presented maximum sizes in the micrometric range is indicative of the
are those for 2 h of milling.
more elongated morphology of the particles (nanofibrils – Fig. 6)
[7,47]. This size trend can be observed in Fig. 5, which correlates the
3.2. Preliminary evaluation of sizes and shapes and morphological analysis average particle size and ultrasound time.
Correlations with the SLS technique and morphological analyses
The milling was conducted in four different times: 1, 2, 3, and 4 h, were needed to confirm the shape of the NCs. In this way, SLS analysis
and the effects of the different times were evaluated dimensionally. The was conducted, and Table 3 presents the obtained values of the hy-
DLS and SLS analysis allow a preliminary evaluation of the NC’ sizes drodynamic radius (RH), rotation radius (RG), distribution of sizes, and
and shapes. The DLS measures the translational diffusion coefficient of identification of the geometric shape of objects (RG/RH).
particles in water, which undergo Brownian motion. The value given by With the SLS data, it is possible to evaluate the parameter ρ, related
5
R.R. Ferreira, et al. Materials Today Communications 22 (2020) 100755
Fig. 3. Challenges of ball mill in the nanocellulose (NC) isolation, comparison between the real and the aimed scenario, and the main techniques that can be
employed to analyze each NC characteristic.
to the RG (obtained from SLS values) and RH (obtained from DLS va- (ρ = 1.48) and 2 M-20U (ρ = 1.22) are considered as an excellent
lues). The RG is a parameter defined as an average root mean squared morphology due to its ρ, which fits into the fibrillar morphology be-
distance from the center of the mass of a macromolecule. This para- cause they present the ratio of ρ > 1 [48].
meter allows a correlation between the RG/RH and the morphology of To corroborate and complement the DLS and SLS results, the mor-
the systems studied. It is a little-used technique for understanding cel- phological analysis was conducted. Fig. 6a shows the FESEM images of
lulosic systems because of its complexity; however, it provides basic the milled and ultrasonicated samples. The FESEM images validate the
information in a short time. changes in the disintegration of the fibers, and the best samples were
According to Stetefeld et al., RG/RH ˜1 can be attributed to elon- evaluated by TEM images, as shown in Fig. 6b that is the 2 h milled
gated shapes [48]. It is possible to observe that there is much different sample and 2 M assisted by 20 min of high-intensity ultrasound [70].
information that needs to be compilated and corroborated with other The best samples were determined according to the previous results
morphological techniques. In the case of NCs, many shapes can occur, presented in Fig. 4S.
such as spheres, fibrils, and rod-like structures [31,49–51]. So, there is a The use of FESEM is limitated by the equipment conditions, being
limitation in the precise estimation of the shapes, resulting in approx- necessary to dry the samples, which results in agglomeration and
imate values. overestimation of sizes. NC usually aggregate into bundles during the
According to the values, the time of 20 and 30-min ultrasound are drying process, and that is why FESEM images are not so easy to see and
considered adequate to obtain an elongated shape. The sample 1 M-30U individual nanoparticles could not be observed. Also, the surface of the
6
R.R. Ferreira, et al. Materials Today Communications 22 (2020) 100755
1M values. Considering the δ values of the samples (Table 3), after the
2M ultrasound, the samples 4 M reduced the stability of the suspension, for
400 3M
3 M's there is no significant change in this properties, and the 2 M
4M
samples increased the stability for longer times of ultrasound, whereas
the 1 M’s the same behavior was observed. The sample 2 M-20U stood
out with ˜60 mV, which means that there is possibly a greater avail-
300
ability of free hydroxyls that result in repulsive electrostatic charges
Size (nm)
7
R.R. Ferreira, et al. Materials Today Communications 22 (2020) 100755
Fig. 6. (a) FESEM images of the milled and ultrasonicate samples and (b) TEM micrographs of the best conditions: 2 M and 2 M-20U; (c) Scheme shows the main
consideration about the effect of the high-intensity ultrasonic.
times evaluated, the use of 20 min presented more uniform sizes and 3.5. Evaluation of the composition and eventual contamination of the NCs
smaller size distribution. Thus, the most effective combination of
methods was for 2 h of milling assisted with 20 min of ultrasound. The high-resolution XPS spectra were collected for the milled
samples and the ultrasonicated samples for 20 min (Fig. 8c–d), re-
spectively. It was observed that increasing the milling time there is the
presence of Al and Si. These elements are impurities associated with the
8
R.R. Ferreira, et al. Materials Today Communications 22 (2020) 100755
Table 3 milled samples. This difference in NC values is due to the chemical pre-
Structural parameters using light scattering techniques (RH values were ob- treatment and the combinatorial isolation method used to obtain NCs
tained from DLS and RG values from SLS), and zeta potential values. from the ES fibers, in which the ball mill allowed the partial removal of
Samples RH (nm) RG (nm) ρ (RG/RH) Zeta potential (mV) oxygen from the NCs while the high-intensity ultrasound caused a
closer increase of pure NC (0.83), as reported in the literature [62].
1M 2840.0 338.3 0.12 −24.7
1 M-10U 453.6 132.6 0.29 −51.7
1 M-20U 361.4 156.6 0.43 −40.8
3.6. Thermogravimetric analysis (TGA)
1 M-30U 113.4 167.3 1.48 −40.9
2M 246.8 221.4 0.90 −43.5
2 M-10U 450.9 173.5 0.39 −42.1 The thermal analysis gives information about the thermal behavior
2 M-20U 145.9 177.3 1.22 −59.3 of the ES and NC derived from it as well as possible high-temperatures
2 M-30U 826.0 150.8 0.18 −53.5 applications of NC [22]. The results obtained by the thermogravimetric
3M 2600.0 306.8 0.19 −45.0
3 M-10U 544.6 197.5 0.36 −45.7
analysis were plotted as thermogravimetric and its derivative curves, as
3 M-20U 488.8 156.1 0.32 −39.3 in Fig. 9a–b.
3 M-30U 277.6 167.7 0.60 −45.7 The TGA curves maintained a similar pattern for all the samples,
4M 240.1 207.3 0.86 −45.7 which is in agreement with the conventional curves of NC [53,63–68].
4 M-10U 301.4 131.1 0.44 −48.1
All these samples showed two main events of percentage weight
4 M-20U 374.5 143.4 0.38 −40.7
4 M-30U 216.8 154.3 0.71 −40.5 loss. The first is related to the breakage of inter- and intramolecular
hydrogen bonds corresponding to water evaporation [30,58]. The
second event of weight loss is related to the thermal degradation of
jar and balls composition (SiO2 and alumina), and higher times of cellulose. The ES sample had slightly higher thermal degradation
milling resulted in higher contents of impurities. The O/C ratio was temperatures than the NCs. Nanoparticles have larger surface areas
calculated, and the obtained values are 0.39, 0.49, 0.54, 0.68, and 0.75 than ES, which may have catalyzed thermal degradation. All the milled
for the samples ES, 1 M, 2 M, 3 M, and 4 M, respectively. The theoretical samples showed very similar Tmax, even at high mechanical milling
value of pure cellulose is 0.83 [61]. times, which indicates that all the structures are identical, and similar
Alumina is the main composition of the jar and spheres; Silicon energies are required to thermal degradation of the cellulose.
dioxide is the "contaminant" mostly found in the balls. Interestingly, as There was an increase in decomposition temperature by 10 °C in all
milling time increased, the aluminum concentration was constant and the ultrasonically treated samples than ball-milled samples. This small
low; however, there was an increase in silicon content due to the wear change could be due to the more ordered and individualized structures
of the balls, which mainly occurs in the region of impurities, resulting which corroborate with the previously presented crystallinity results.
in contamination of the cellulose. The use of other materials as grinding Another possibility is that the surface area of the NC is very high, which
agents such as high alumina or zirconic spheres could significantly re- means that the heat can be disseminated very quick, and it can decrease
duce contamination but would result in an increase of approximately 20 the thermal decomposition temperature of the samples.
% in cost. This behavior is another indication of the efficiency of the combined
For the ultrasonicated samples, the values are 0.61, 0.64, 0.61, and methods since the high temperatures are justified by the higher energy
0.79 for the samples 1 M-20U, 2 M-20U, 3 M-20U, and 4 M-20U, re- necessary for the thermal degradation of the nanocelluloses [69]. These
spectively. The ultrasound samples showed higher O/C values than the results are promising and demonstrate that the combination of the two
insulation methods may favor applications that require higher
9
R.R. Ferreira, et al. Materials Today Communications 22 (2020) 100755
Fig. 8. X-ray diffraction patterns of (a) milled samples and (b) ultrasonicated samples; XPS survey spectra (0–1400 eV) for (c) eucalyptus sawdust and milled NCs and
(d) ultrasonicated samples, showing the representative C 1s (286 eV) and O 1s (532 eV) photoemission peaks.
10
R.R. Ferreira, et al. Materials Today Communications 22 (2020) 100755
Fig. 9. Thermal behavior of the ES sample and their prepared nanoparticles. ES and their nanocelluloses thermograms without (a) and with (b) the use of ultrasound.
The inserts are the corresponding DTG curves.
11
R.R. Ferreira, et al. Materials Today Communications 22 (2020) 100755
https://doi.org/10.1016/j.carbpol.2016.04.076. 10.1021/bk-2017-1251.ch004.
[9] S. Bee, A. Hamid, H.Vc. of L.B. to N.: S. and C.P.C. Lee, S.B. a Hamid, S.K. Zain, rsion [34] U.P. Agarwal, Raman Spectroscopy of CNC- and CNF-based nanocomposites,
of Lignocellulosic Biomass to Nanocellulose :, 2014 (2015). doi:10.1155/2014/ Handb. Nanocellulose Cellul. Nanocompos. (2017) 609–625, https://doi.org/10.
631013. 1002/9783527689972.ch18.
[10] A.G. de Souza, D.B. Rocha, F.S. Kano, D.dosS. Rosa, Valorization of industrial paper [35] A. Jäger, E. Jäger, F. Surman, A. Höcherl, B. Angelov, K. Ulbrich, M. Drechsler,
waste by isolating cellulose nanostructures with different pretreatment methods, V.M. Garamus, C. Rodriguez-Emmenegger, F. Nallet, P. Štěpánek, Nanoparticles of
Resour. Conserv. Recycl. 143 (2019) 133–142, https://doi.org/10.1016/j. the poly([N-(2-hydroxypropyl)]methacrylamide)-b-poly[2-(diisopropylamino)ethyl
resconrec.2018.12.031. methacrylate] diblock copolymer for pH-triggered release of paclitaxel, Polym.
[11] W. Pan, X. Zhang, F. Niu, M. Li, J. Li, J. Yang, Q. Huang, The characteristic and Chem. 6 (2015) 4946–4954, https://doi.org/10.1039/c5py00567a.
dispersion stability of nanocellulose produced by mixed acid hydrolysis and ultra- [36] H.L. Yi, L.J. Lai, J.S. Jiang, C.C. Hua, Ethylcellulose colloids incubated in dilute
sonic assistance, Carbohydr. Polym. 165 (2017) 197–204, https://doi.org/10.1016/ solution, J. Phys. Chem. B 121 (2017) 638–648, https://doi.org/10.1021/acs.jpcb.
j.carbpol.2017.02.048. 6b09976.
[12] Z. Karim, S. Afrin, Q. Husain, R. Danish, Necessity of enzymatic hydrolysis for [37] S. Nam, A.D. French, B.D. Condon, M. Concha, Segal crystallinity index revisited by
production and functionalization of nanocelluloses, Crit. Rev. Biotechnol. 37 (2017) the simulation of X-ray diffraction patterns of cotton cellulose Iβ and cellulose II,
355–370, https://doi.org/10.3109/07388551.2016.1163322. Carbohydr. Polym. 135 (2016) 1–9, https://doi.org/10.1016/j.carbpol.2015.08.
[13] C.J. Garvey, P. Memmott, P.K. Annamalai, D.J. Martin, E. Jiang, N. Amiralian, High 035.
aspect ratio nanocellulose from an extremophile spinifex grass by controlled acid [38] L. Segal, J.J. Creely, A.E. Martin, C.M. Conrad, An empirical method for estimating
hydrolysis, Cellulose 24 (2017) 3753–3766, https://doi.org/10.1007/s10570-017- the degree of crystallinity of native cellulose using the X-ray diffractometer, Text.
1379-6. Res. J. 29 (1959) 786–794, https://doi.org/10.1177/004051755902901003.
[14] T. Theivasanthi, F.L. Anne Christma, A.J. Toyin, S.C.B. Gopinath, R. Ravichandran, [39] R.M. de Sá, C.S. de Miranda, N.M. José, Preparation and characterization of na-
Synthesis and characterization of cotton fiber-based nanocellulose, Int. J. Biol. nowhiskers cellulose from fiber arrowroot (Maranta arundinacea), Mater. Res. 18
Macromol. 109 (2018) 832–836, https://doi.org/10.1016/j.ijbiomac.2017.11.054. (2015) 225–229, https://doi.org/10.1590/1516-1439.366214.
[15] J. Hu, D. Tian, S. Renneckar, J.N. Saddler, Enzyme mediated nanofibrillation of [40] R.A. Ilyas, S.M. Sapuan, M.R. Ishak, Isolation and characterization of nanocrystal-
cellulose by the synergistic actions of an endoglucanase, lytic polysaccharide line cellulose from sugar palm fibres (Arenga Pinnata), Carbohydr. Polym. (2017),
monooxygenase (LPMO) and xylanase, Sci. Rep. 8 (2018) 1–15, https://doi.org/10. https://doi.org/10.1016/j.carbpol.2017.11.045.
1038/s41598-018-21016-6. [41] F. Kallel, F. Bettaieb, R. Khiari, A. García, J. Bras, S.E. Chaabouni, Isolation and
[16] N. Kruer-Zerhusen, B. Cantero-Tubilla, D.B. Wilson, Characterization of cellulose structural characterization of cellulose nanocrystals extracted from garlic straw
crystallinity after enzymatic treatment using Fourier transform infrared spectro- residues, Ind. Crops Prod. 87 (2016) 287–296, https://doi.org/10.1016/j.indcrop.
scopy (FTIR), Cellulose 25 (2018) 37–48, https://doi.org/10.1007/s10570-017- 2016.04.060.
1542-0. [42] J.S. Lupoi, E. Gjersing, M.F. Davis, Evaluating lignocellulosic biomass, its deriva-
[17] M. Nuruddin, M. Hosur, M. Jamal Uddin, D. Baah, S. Jeelani, A novel approach for tives, and downstream products with Raman spectroscopy, Front. Bioeng.
extracting cellulose nanofibers from lignocellulosic biomass by ball milling com- Biotechnol. 3 (2015) 1–18, https://doi.org/10.3389/fbioe.2015.00050.
bined with chemical treatment, J. Appl. Polym. Sci. 133 (2016), https://doi.org/10. [43] Y. Boluk, C. Danumah, Analysis of cellulose nanocrystal rod lengths by dynamic
1002/app.42990. light scattering and electron microscopy, J. Nanopart. Res. 16 (2014), https://doi.
[18] M.C. Guimara, Evaluation of the Effect of Processing Conditions on the Impact org/10.1007/s11051-013-2174-4.
Properties of, (2017), https://doi.org/10.1177/0021998316688338. [44] Y. Mao, K. Liu, C. Zhan, L. Geng, B. Chu, B.S. Hsiao, Characterization of nano-
[19] S. Mondal, Preparation, properties and applications of nanocellulosic materials, cellulose using small-angle neutron, X-ray, and dynamic light scattering techniques,
Carbohydr. Polym. 163 (2017) 301–316, https://doi.org/10.1016/j.carbpol.2016. J. Phys. Chem. B 121 (2017) 1340–1351, https://doi.org/10.1021/acs.jpcb.
12.050. 6b11425.
[20] F.S. Kano, A.G. de Souza, D.dosS. Rosa, Variation of the milling conditions in the [45] V.K. Baheti, R. Abbasi, J. Militky, Ball milling of jute fibre wastes to prepare na-
obtaining of nanocellulose from the paper sludge, Matéria (Rio Janeiro). 24 (2019), nocellulose, World J. Eng. 9 (2012) 45–50.
https://doi.org/10.1590/s1517-707620190003.0719. [46] M.R.K. Sofla, R.J. Brown, T. Tsuzuki, T.J. Rainey, A comparison of cellulose na-
[21] C.C. Piras, S. Fernández-Prieto, W.M. De Borggraeve, Ball milling: a green tech- nocrystals and cellulose nanofibres extracted from bagasse using acid and ball
nology for the preparation and functionalisation of nanocellulose derivatives, milling methods, Adv. Nat. Sci. Nanosci. Nanotechnol. 7 (2016), https://doi.org/
Nanoscale Adv. 1 (2019) 937–947, https://doi.org/10.1039/c8na00238j. 10.1088/2043-6262/7/3/035004.
[22] C.M.E. Xuran, L. Min, W.Y. Huang, C.M. Ewulonu, Á.X. Liu, Á.M. Wu, Á.Y. Huang, [47] A.N. Frone, D.M. Panaitescu, I. Chiulan, C.A. Nicolae, Z. Vuluga, C. Vitelaru,
X. Liu, Ultrasound-assisted mild sulphuric acid ball milling preparation of lig- C.M. Damian, The effect of cellulose nanofibers on the crystallinity and nanos-
nocellulose nanofibers (LCNFs) from sunflower stalks (SFS), Cellulose 4 (2019), tructure of poly(lactic acid) composites, J. Mater. Sci. 51 (2016) 9771–9791,
https://doi.org/10.1007/s10570-019-02382-4. https://doi.org/10.1007/s10853-016-0212-1.
[23] Z. Ling, T. Wang, M. Makarem, M. Santiago Cintrón, H.N. Cheng, X. Kang, [48] J. Stetefeld, S.A. McKenna, T.R. Patel, Dynamic light scattering: a practical guide
M. Bacher, A. Potthast, T. Rosenau, H. King, C.D. Delhom, S. Nam, J. Vincent and applications in biomedical sciences, Biophys. Rev. 8 (2016) 409–427, https://
Edwards, S.H. Kim, F. Xu, A.D. French, Effects of ball milling on the structure of doi.org/10.1007/s12551-016-0218-6.
cotton cellulose, Cellulose 26 (2019) 305–328, https://doi.org/10.1007/s10570- [49] T.T.T. Ho, T. Zimmermann, R. Hauert, W. Caseri, Preparation and characterization
018-02230-x. of cationic nanofibrillated cellulose from etherification and high-shear disintegra-
[24] Y. Zheng, Z. Fu, D. Li, M. Wu, Effects of ball milling processes on the microstructure tion processes, Cellulose 18 (2011) 1391–1406, https://doi.org/10.1007/s10570-
and rheological properties of microcrystalline cellulose as a sustainable polymer 011-9591-2.
additive, Materials (Basel) 11 (2018) 1–13, https://doi.org/10.3390/ma11071057. [50] L. Zhang, T. Tsuzuki, X. Wang, Preparation of cellulose nanofiber from softwood
[25] H.P.S. Abdul Khalil, Y. Davoudpour, M.N. Islam, A. Mustapha, K. Sudesh, pulp by ball milling, Cellulose 22 (2015) 1729–1741, https://doi.org/10.1007/
R. Dungani, M. Jawaid, Production and modification of nanofibrillated cellulose s10570-015-0582-6.
using various mechanical processes: a review, Carbohydr. Polym. 99 (2014) [51] R. Prathapan, R. Thapa, G. Garnier, R.F. Tabor, Modulating the zeta potential of
649–665, https://doi.org/10.1016/j.carbpol.2013.08.069. cellulose nanocrystals using salts and surfactants, Colloids Surf. A Physicochem.
[26] A. Barakat, C. Mayer-Laigle, A. Solhy, R.A.D. Arancon, H. De Vries, R. Luque, Eng. Asp. 509 (2016) 11–18, https://doi.org/10.1016/j.colsurfa.2016.08.075.
Mechanical pretreatments of lignocellulosic biomass: towards facile and en- [52] Y. Xu, A.D. Atrens, J.R. Stokes, Rheology and microstructure of aqueous suspen-
vironmentally sound technologies for biofuels production, RSC Adv. 4 (2014) sions of nanocrystalline cellulose rods, J. Colloid Interface Sci. 496 (2017) 130–140,
48109–48127, https://doi.org/10.1039/c4ra07568d. https://doi.org/10.1016/j.jcis.2017.02.020.
[27] R.J. Moon, A. Martini, J. Nairn, J. Simonsen, J. Youngblood, Cellulose [53] S. Naduparambath, J. T.V, V. Shaniba, S. M.P, A.K. Balan, E. Purushothaman,
Nanomaterials Review: Structure, Properties and Nanocomposites, (2011), https:// Isolation and characterisation of cellulose nanocrystals from sago seed shells,
doi.org/10.1039/c0cs00108b. Carbohydr. Polym. 180 (2018) 13–20, https://doi.org/10.1016/j.carbpol.2017.09.
[28] J. Desmaisons, E. Boutonnet, M. Rueff, A. Dufresne, J. Bras, A new quality index for 088.
benchmarking of different cellulose nanofibrils, Carbohydr. Polym. 174 (2017) [54] A. Pappu, M. Saxena, V.K. Thakur, A. Sharma, Facile extraction, processing and
318–329, https://doi.org/10.1016/j.carbpol.2017.06.032. characterization of biorenewable sisal fibers for multifunctional applications, J.
[29] M.S. Reid, M. Villalobos, E.D. Cranston, Benchmarking cellulose nanocrystals: from Macromol. Sci. Part A 1325 (2016), https://doi.org/10.1080/10601325.2016.
the laboratory to industrial production, Langmuir 33 (2017) 1583–1598, https:// 1176443.
doi.org/10.1021/acs.langmuir.6b03765. [55] A.C. Corrêa, E.M. Teixeira, L.A. Pessan, L.H.C. Mattoso, Cellulose nanofibers from
[30] G.F. de Lima, A.G. de Souza, D.S. Rosa, Effect of adsorption of polyethylene glycol curaua fibers, Cellulose 17 (2010) 1183–1192, https://doi.org/10.1007/s10570-
(PEG), in aqueous media, to improve cellulose nanostructures stability, J. Mol. Liq. 010-9453-3.
268 (2018) 415–424, https://doi.org/10.1016/j.molliq.2018.07.080. [56] C. Brinatti, J. Huang, R.M. Berry, K.C. Tam, W. Loh, Structural and energetic studies
[31] A.G. De Souza, F.S. Kano, J.J. Bonvent, D.S. Rosa, Cellulose Nanostructures on the interaction of cationic surfactants and cellulose nanocrystals, Langmuir 32
Obtained From Waste Paper Industry : A Comparison of Acid and Mechanical (2016) 689–698, https://doi.org/10.1021/acs.langmuir.5b03893.
Isolation Methods, (2017), pp. 1–6. [57] A. Kaboorani, B. Riedl, Surface modification of cellulose nanocrystals (CNC) by a
[32] C. Lee, K. Dazen, K. Kafle, A. Moore, D.K. Kohnson, S. Park, S.H. Kim, Correlations cationic surfactant, Ind. Crops Prod. 65 (2015) 45–55, https://doi.org/10.1016/j.
of Apparent Cellulose Crystallinity Determined by XRD, NMR, IR, Raman, and SFG indcrop.2014.11.027.
Methods, in: Adv. Polym. Sci. Springer International Publishing, Switzerland, 2015, [58] H. Lu, Y. Gui, L. Zheng, X. Liu, Morphological, crystalline, thermal and physico-
pp. 1–34, https://doi.org/10.1007/12. chemical properties of cellulose nanocrystals obtained from sweet potato residue,
[33] U.P. Agarwal, Raman spectroscopy in the analysis of cellulose nanomaterials, Food Res. Int. J. 50 (2013) 121–128, https://doi.org/10.1016/j.foodres.2012.10.
Nanocelluloses Their Prep. Prop. Appl. ACS Sympos (2017) 75–90, https://doi.org/ 013.
12
R.R. Ferreira, et al. Materials Today Communications 22 (2020) 100755
[59] W. Chen, H. Yu, Y. Liu, P. Chen, M. Zhang, Y. Hai, Individualization of cellulose coconut husk fibers: Effect of preparation conditions on their thermal and mor-
nanofibers from wood using high-intensity ultrasonication combined with chemical phological behavior, Carbohydr. Polym. 81 (2010) 83–92, https://doi.org/10.
pretreatments, Carbohydr. Polym. 83 (2011) 1804–1811, https://doi.org/10.1016/ 1016/j.carbpol.2010.01.059.
j.carbpol.2010.10.040. [68] S. Shankar, J.W. Rhim, Preparation of nanocellulose from micro-crystalline cellu-
[60] R. Avolio, I. Bonadies, D. Capitani, M.E. Errico, G. Gentile, M. Avella, A multi- lose: The effect on the performance and properties of agar-based composite films,
technique approach to assess the effect of ball milling on cellulose, Carbohydr. Carbohydr. Polym. 135 (2016) 18–26, https://doi.org/10.1016/j.carbpol.2015.08.
Polym. 87 (2012) 265–273, https://doi.org/10.1016/j.carbpol.2011.07.047. 082.
[61] M.E. Achaby, Z. Kassab, A. Barakat, A. Aboulkas, Alfa fibers as viable sustainable [69] T.F. Meyabadi, F. Dadashian, G. Mir Mohamad Sadeghi, H. Ebrahimi Zanjani Asl,
source for cellulose nanocrystals extraction: application for improving the tensile Spherical cellulose nanoparticles preparation from waste cotton using a green
properties of biopolymer nanocomposite films, Ind. CropsProd. 112 (2018) method, Powder Technol. 261 (2014) 232–240, https://doi.org/10.1016/j.powtec.
499–510, https://doi.org/10.1016/j.indcrop.2017.12.049. 2014.04.039.
[62] S. Barazzouk, C. Daneault, Amino acid and peptide immobilization on oxidized [70] J.C. Chandra, N. George, S.K. Narayanankutty, Isolation and characterization of
nanocellulose: spectroscopic characterization, Nanomaterials 2 (2012) 187–205, cellulose nanofibrils from arecanut husk fibre, Carbohydr. Polym. 142 (2016)
https://doi.org/10.3390/nano2020187. 158–166, https://doi.org/10.1016/j.carbpol.2016.01.015.
[63] E. Espinosa, J. Domínguez-Robles, R. Sánchez, Q. Tarrés, A. Rodríguez, The effect of [71] C. Liu, B. Li, H. Du, D. Lv, Y. Zhang, G. Yu, X. Mu, H. Peng, Properties of nano-
pre-treatment on the production of lignocellulosic nanofibers and their application cellulose isolated from corncob residue using sulfuric acid, formic acid, oxidative
as a reinforcing agent in paper, Cellulose 24 (2017) 2605–2618, https://doi.org/10. and mechanical methods, Carbohydr. Polym. 151 (2016) 716–724, https://doi.org/
1007/s10570-017-1281-2. 10.1016/j.carbpol.2016.06.025.
[64] C. Xu, S. Zhu, C. Xing, D. Li, N. Zhu, H. Zhou, Isolation and properties of cellulose [72] G. Chauve, J. Bras, Industrial Point of View of Nanocellulose Materials and Their
nanofibrils from coconut palm petioles by different mechanical process, PLoS One Possible Applications, Https://Umaine.Edu/Pdc/Cellulose-Nanomaterials/Order-
10 (2015) 1–11, https://doi.org/10.1371/journal.pone.0122123. Nanocellulose/, (2014), pp. 233–252, https://doi.org/10.1142/
[65] N. Rambabu, S. Panthapulakkal, M. Sain, A.K. Dalai, Production of nanocellulose 97898145664690014.
fibers from pinecone biomass: Evaluation and optimization of chemical and me- [73] T.P.D.C. of the U. of Maine, Order nanocellulose, Https://Umaine.Edu/Pdc/
chanical treatment conditions on mechanical properties of nanocellulose films, Ind. Cellulose-Nanomaterials/Order-Nanocellulose/, (2018), pp. 4–5 https://umaine.
Crops Prod. 83 (2016) 746–754, https://doi.org/10.1016/j.indcrop.2015.11.083. edu/pdc/cellulose-nano-crystals/.
[66] Y.M. Zhou, S.Y. Fu, L.M. Zheng, H.Y. Zhan, Effect of nanocellulose isolation tech- [74] J. Miller, Nanocellulose State of The Industry, (2015).
niques on the formation of reinforced poly(vinyl alcohol) nanocomposite films, [75] F. Piccinno, R. Hischier, S. Seeger, C. Som, Predicting the environmental impact of a
Express Polym. Lett. 6 (2012) 794–804, https://doi.org/10.3144/expresspolymlett. future nanocellulose production at industrial scale: application of the life cycle
2012.85. assessment scale-up framework, J. Clean. Prod. 174 (2018) 283–295, https://doi.
[67] M.F. Rosa, E.S. Medeiros, J.A. Malmonge, K.S. Gregorski, D.F. Wood, org/10.1016/j.jclepro.2017.10.226.
L.H.C. Mattoso, G. Glenn, W.J. Orts, S.H. Imam, Cellulose nanowhiskers from
13